The unprecedented use of computers in all areas of human activity makes the challenge of building errorless programs the central one of informatics.
The challenge is of utmost importance for developing built in program systems for managing critical applications where defects of programming can lead to catastrophic consequences. The cost of activities to ensure errorless behavior of built in software management systems amounts to more than a half of their development cost.
Modern day methods of software quality assurance include a variety of means, techniques, and approaches. Although testing and debugging stay the main method, verification, i.e. the formal proof of meeting a set of formal requirements by a formal program system model, has been gaining wider use recently. The method of Model checking has become a breakthrough trend in the area of formal verification.
The method of Model checking can be used for software and hardware systems that represent a model of transformations with a finite number of states. Therefore, the main problem of this method is the “state explosion problem”, i.e. the exponential growth of the number of states of a parallel programs system as the number of interacting components grows. The development of “symbol” algorithms based on economical methods of representing the final data structures has lead to a reduction of this method’s sensitivity to the “state explosion problem” and a significant increase in the efficiency of this method of verification. The “symbol” methods of verification have been successfully used in many practical development projects of building real program systems. Currently this technique is used as a technological phase in many large firms who develop built in systems for critical applications.
Model checking stays a “hot” area of informatics, as intensive research continues to be underway and as the means of this approach are being broadened.
A group of leading scientists in the area of applying formal software development methods has launched an ambitions international research project named “Verified Software Initiative”. The goal of the project is to bring the theoretical foundation, the instruments, and the elements of the verification technology to a state where they will allow developing errorless software systems.